Richard> What is the status of the SpamBayes project? I really love it
Richard> and I recommend it to everyone I talk to but it seems like it's
Richard> getting stale. Are you guys still maintaining it or should I
Richard> look elsewhere for my spam protection? :)
Is it failing as a result of being stale? My experience is that mold has a
hard time getting a good grip on electrons.
As Bob Coe indicated, the algorithm is sound. The composition of your
training database is key to keeping things running smoothly. I fairly
routinely delete old messages from my database, especially old spam. I
think that helps in a few ways:
* keeps the database from growing without bound
* purges the database of mistakes
* purges the database of messages that no longer represent the sort of
mail I get now
That said, I'm not an Outlook user. I have no idea how to purge old
messages in that environment. If you suspect problems with your database,
clearing it and starting from scratch my be the best route.
Skip
_______________________________________________
[email protected]
http://mail.python.org/mailman/listinfo/spambayes
Check the FAQ before asking: http://spambayes.sf.net/faq.html
